[Amps] High power attenuators (HPA's)

[EVonvaltie@aol.com]

I have some interesting info to share with other engineering minded hams. I 
have long been plagued by the problem of trying to make accurate measurments 
at high power levels. A simple example is  the calibration of a wattmeter. I 
have never seen a device on the market which fills the bill for this need, 
especially at a decent price. So I have finally taken the time time and 
effort to do it myself.

I just ran off the final phase and amplitude plots of the first finished 
model of a new type of attenuator, which produces 70 db atten. +/- .5 db and 
+/-3 degrees phase shift across the band from 2-30 MHz. It also works at 
30-150MHZ with a small but known error in amplitude/phase. I have built it up 
to emulate the same form factor as a typical inline precision attenuator( 
e.g. HP,Weinschel,etc.) with an N-male input and BNC-female 50 ohm output. It 
is not a true attenuator - that is a power attenuator. It is a voltage 
attenuator which minimally loads the measured node but produces an accurate 
voltage output. For example, 1kW at the input connector results in -10dbm 
into 50 ohms at the output, which is a nice fit for analyzers, vector 
voltmeters, etc. 100 watt transmitters come out at -20dbm, etc. It is 
capacitively coupled and dissipates essentially no heat.

The construction is too complicated for  normal do-it-yourselfers, with the 
key element being a specially designed miniature high-voltage coaxial 
capacitor. I am considering marketing it sometime later this year (if and 
when I get caught up with all current business.) Meanwhile, I would like to 
build up a small batch of them for friends,who can also give me some feedback 
on the finished unit.I do not want to go into any overtly commercial 
overtures here, as I believe in keeping our hobby as uncommercial as is 
feasible. My market for this device would more likely center on the 
industrial RF world.

If you are interested,  contact me directly and I will discuss in more detail.

I will not be at Dayton personally, but I may make arrangements to send a 
working model with a friend of mine, who can show it to interested users. If 
that becomes QRV, I will post it on here on Thursday.

Eric von Valtier K8LV


--- StripMime Report -- processed MIME parts ---
multipart/alternative
  text/plain (text body -- kept)
  text/html
---